Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Tone recognition in continuous Cantonese speech using supratone models.

Yao Qian1, Tan Lee, Frank K Soong

  • 1Department of Electronic Engineering, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong.

The Journal of the Acoustical Society of America
|June 7, 2007
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Tissue Heterogeneity-Driven Parallel Acquisition for High-Coverage MS/MS Imaging of Lipids.

Angewandte Chemie (International ed. in English)·2026
Same author

Priming iTBS for lower limb rehabilitation after stroke: Protocol for a randomized controlled trial on efficacy and neuroplasticity.

PloS one·2026
Same author

Effectiveness of Blood Flow Restriction Training in Neurological Rehabilitation: A Systematic Review and Meta-analysis.

Sports medicine - open·2026
Same author

An investigation into rail corrugation evaluation based on fractal theory and explicit finite element method.

Scientific reports·2026
Same author

BPHL promotes TNBC stemness by resolving R-loops via POLR2A lactylation inhibition and BARD1-mediated ubiquitination.

Cancer letters·2026
Same author

Surgically Oriented Ultrasound-Based AI of Median Nerve Morphology as a Decision Support for Carpal Tunnel Release: A Calibrated ConvNeXt-CBAM Framework With Supervised Contrastive Warm-Up.

Annali italiani di chirurgia·2026

This study introduces supratone modeling for Cantonese tone recognition, improving accuracy by analyzing pitch contours across multiple tones. The method enhances automatic speech recognition for this tonal language.

Area of Science:

  • Computational Linguistics
  • Speech Processing
  • Phonetics

Background:

  • Cantonese, a major Chinese dialect, possesses a rich tonal system crucial for meaning.
  • Automatic speech recognition (ASR) for Cantonese benefits from accurate tone recognition.
  • Challenges in Cantonese tone recognition include similar pitch contour shapes and overlapping F0 ranges.

Purpose of the Study:

  • To develop a novel method for automatic tone recognition in continuous Cantonese speech.
  • To address the acoustic indistinguishability of tones within syllables due to pitch variations.
  • To improve the accuracy of Cantonese tone recognition, particularly for tones with similar contours.

Main Methods:

  • Proposed a supratone modeling approach, characterizing the fundamental frequency (F0) contour of successive tones.

Related Experiment Videos

  • Modeled tone sequences as overlapped concatenations of supratone units.
  • Incorporated phonological constraints on syllable-tone combinations to determine the most likely tone sequence.

    • Main Results:

    • Achieved 74.68% accuracy in speaker-independent Cantonese tone recognition experiments.
    • Demonstrated that relative pitch heights between neighboring tones are largely preserved.
    • Significantly reduced confusion among Cantonese tones with similar contour shapes.

    Conclusions:

    • Supratone modeling effectively captures inter-tone pitch relationships, enhancing Cantonese tone recognition.
    • The proposed method offers a robust solution for a challenging aspect of Cantonese speech processing.
    • This work contributes to advancing automatic speech recognition capabilities for tonal languages.